Experimental Physics and
Industrial Control System

Maren Purves <[email protected]> · Wed, 27 Jun 2007 12:09:31 -1000

<g>
working in an environment that has been in operation for over 25 years
(with bits and pieces removed and added every so often), a short
and a bit facetious answer is: very carefully :-)
(and I know we are in a much better situtation than e.g. Matthias,
who told me he sometimes can't reboot an ioc for over a year)
Our VAX system (we still have VAXes, and we use them) is frozen at
VMS 5.5.2.
We still have a solaris machine so that we can keep ar_cmd running
and that we can still compile transputer software.
We don't upgrade things like EPICS unless we have scheduled downtime
(so we're still running 3.13.8)
We have a development environment that's in a different location
(where the office is rather than where the telescopes are).
Sometimes we have hardware to work with, sometimes we don't.
When we don't we work with simulators and hope that moving to
real hardware won't break things. Sometimes that happens and
sometimes it doesn't. I think the real answer to having only one
piece of hardware is simulators. Whenever you get access to the
hardware you use it, check how far you've gotten this time, and
move back to the simulator environment when they take the hardware
away again (this can happen for other reasons than actual real
operations, like when your detector gets sent off for quantum
efficiency measurements - been there, done that)
In our group my machine is the only one that still runs WBEL4 Linux
as I don't have the time to deal with an upgrade, everybody else
runs CentOS 5. But I'm talking about people's desktop machines here
from which we, if need be, log into wherever to do the actual work.
Virtual machines may work if you can emulte the environment that you
need. It doesn't always work.
Hope that answers at least some questions.
Aloha,
Maren
(who sometimes tries to do some development while still doing lots
of operational support, as in 'whenever something breaks')

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System